vreg.Volume3D.find_transform_rigid_to#

Volume3D.find_transform_rigid_to(target, params=None, metric=None, optimizer=None, resolutions=None, mask=None, target_mask=None, center=None, coords='fixed')[source]#

Find the rigid transform to a static target volume.

Parameters:

  • target (vreg.Volume3D) – Fixed target volume for the coregistration.

  • params (array-like) – 6-element vector with translation and rotation vectors, in that order.

  • metric (func, optional) – Metric to quantify goodness of alignment. Options are ‘mi’ (mutual information), ‘sos’ (sum of squares), and ‘migrad’ (mutual information of the image gradient). Default is ‘mi’.

  • optimizer (dict, optional) – Optimizer as a dictionary with one key method specifying the method used for optimization. The other items in the dictionary are any optional keywords accepted by the method. Defaults to {‘method’:’LS’}.

  • resolutions (list, optional) – Resolutions to use in the optimization. Defaults to [1].

  • mask (Volume3D, optional) – volume used for masking the moving volume. Defaults to None.

  • target_mask (Volume3D, optional) – volume used for masking the static volume. Defaults to None.

  • center (str or array-like) – center of rotation. If this has value ‘com’, the rotation is performed around the center of mass. Alternatively this can be a 3-element vector. Defaults to None (center = origin).

  • coords (str or array, optional) – Reference frame for the coordinates of the translation, rotation and center vector, as a 4x4 affine array. String options are shorthand notations: ‘fixed’ (patient reference frame), and ‘volume’ (volume reference frame). Defaults to ‘fixed’.

Returns:

The optimal values for the transformaton parameters.

Return type:

params